Hardware Implementation of Connected Components Analysis for the Vision-based Tactile Perception Image

In this paper, a novel connected components analysis algorithm of the vision-based tactile perception image was presented. There aren't complex connected components after morphological filtering in these images. So there is no need to resolve label equivalence problem when performing connected components analysis. In the hardware implementation of the proposed algorithm, the memory requirement was reduced by more than 41%. The speed of image processing is twice as fast as that of the classical algorithm.